Attainment of maximal peak bone mass is crucial in the prevention of osteoporosis. Because a majority of bone mass is accrued during the peri- pubertal years, intervention designed to increase the peak should target this period. Physical activity is an important modifiable factor contributing to bone accrual, yet it is unclear whether benefits accrued during childhood are maintained to skeletal maturity. This proposal tests the following hypotheses: (I) Girls who participate in impact activity will accrue greater amounts of bone during the peripubertal years than inactive girls, and will retain those benefits even after cessation of the activity; (II) Increases in bone geometry will provide a greater contribution to bone mass gain than increases in bone material properties. To test these hypotheses the following Specific Aims are proposed: (I) To longitudinally measure bone accrual and the amount of impact activity in a cohort of pre- and peri-pubertal gymnasts, retired gymnasts, and non- gymnast controls will be followed for a 3 year period, beginning at an average of 11 years. Hours per week of gymnastics activity will be recorded and amount of physical activity will be measured by questionnaire. At baseline and six month intervals anthropemetric measurements will be made, self-Tanner stage will be recorded, and a semi-quantitative food frequency questionnaire will be administered. At yearly intervals bioelectric impedance analysis, muscle strength and endurance will be measured, and dual energy x-ray absorptiometry (DXA) scans will be performed. (II) To longitudinally measure bone geometry and bone mineral material properties in the same activity and control cohorts. DXA scans will be used to assess differences in bone geometry and in bone and control cohorts. DXA scans will be used to assess differences in bone geometry and in bone volumetric density activity vs. inactive children. Confirmation of the proposed hypotheses will affirm the importance of childhood activity in bone accrual, providing evidence of the protective effect of such activity.